Re: bonnie and 2.3.8
Catalin Muresan (cata@codec.ro)
Thu, 24 Jun 1999 15:04:10 +0300
On Thu, Jun 24, 1999 at 12:34:40AM +0200, Guest section DW wrote:
dwguest> > 03:01: rw=0, want=2138996093, limit=2015968
dwguest>
dwguest> We might want to add
dwguest>
dwguest> printk(KERN_INFO "dev %s blksize=%d blocknr=%d sector=%d size=%d count=%d\n",
dwguest> kdevname(bh->b_dev),
dwguest> blksize_size[major][MINOR(bh->b_dev)],
dwguest> bh->b_blocknr, bh->b_rsector, bh->b_size, bh->b_count);
dwguest>
dwguest> after the printk(KERN_INFO "%s: rw=%d, want=%d, limit=%d\n", ... b_rdev)]);
dwguest> in make_request() in ll_rw_blk.c - just to make sure that all is OK
dwguest> (no buffer_head corruption) and only b_blocknr is ridiculous.
dwguest>
dwguest> As far as I recall, this kind of messages has mostly been reported by people
dwguest> using FAT, where it may be [have been] a fs problem, and by people who
dwguest> got this error together with other SCSI-related error messages.
I've added the printk:
Jun 24 14:57:10 Cnn kernel: attempt to access beyond end of device
Jun 24 14:57:10 Cnn kernel: 03:01: rw=0, want=656811301, limit=2015968
Jun 24 14:57:10 Cnn kernel: dev 03:01 blksize=1024 blocknr=656811300 sector=1313622600 size=1024 count=1
the numbers look ok - b_blocknr is too large.
this happens with:
- kernel 2.3.9-pre1 (also with 2.3.8-pre2, 2.3.8)
- ide drive (03:01), ext2 root partition (not fat)
[root@Cnn]:bonnie/:# ./bonnie
File './Bonnie.893', size: 104857600
Writing with putc()...done
Rewriting...^C
as soon as bonnie gets to the `rewriting' part the errors appear.
--
Catalin Muresan
CODEC Electronic Products
Internet Services Department
699389.26 632110.85
-
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@vger.rutgers.edu
Please read the FAQ at http://www.tux.org/lkml/